a tm o sp he ric Lrviwm wtr Vol. 21. No. 9. pp. 1891-1898. 1987 Printe d in G re a t Brita in. oMJ4-6981/ %7 s3.oo+o.@.l Pc rg a m o n Jo urna ls Ltd . WIND TUNNEL EXPERIMENTS ON THE RESUSPENSION OF SUB-MICROMETER PARTICLES FROM A SAND SURFACE STEPHAN BORRMANN and RUPRECHT JAENICKE lnstitut fiir Meteorologic. University of Main& D-6500 Maim, F.R.G. ( zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA First receiued 13 May 1986 and receiwd fir publication 6 March 1987) Abstract-In wind tunnel experiments rub-pm aerosol size distributions have been measured within the turbulent boundary layer above a sand surface that has been under the eroding influence of a quantified flow of saltoting glassbeads (50 pm radius). An optical particle counter covering a particle size radius range from 0.33 lo 1.01 pm was used for measuring size distributions as a function of height above the sand surface and shear velocity. Vertical, upward-directed particle fluxes have been calculated under Ihe assumption of a vertical aerosol exchange similar to that of momentum. The measurementsshow a strong dependency of the particle flux on the shear velocity u*. It can be described by a third order polynomial fit of the flux vs u* data. Also the range of particle size radii from 0.71 to 1.01pm seems to contain the lower limit of the sand blasting aerosol sixe.The results of the measurements of this study are compared lo other wind tunnel studiesand to field data as well. Key word index: Atmospheric aerosols. desert as source of aerosols. resuspension experiments in wind tunnels, saltation, sand surface wind erosion. I. INTRODUCTION Desert regions contribute to the production of natural aerosols in quantities comparable to the sea surfaces (Schiitz, 1980). Deserts are stronger sources than volcanic eruptions and other sources. Desert-derived mineral dust particles with particle size radii from 0.1 to 1 pm are of particular interest because they have the longest residence times in the atmosphere compared to smaller and larger aerosol particles (Jaenicke, 1980). and thus are subject to long-range transport. Several authors studied the source strength of the Sahara Desert and modeled the long-range transport (d’Almeida, 1983; Schlitz, 1980). Carlson and Benjamin (1980) estimated the effects of a dust-laden air mass from the Sahara Desert on the tropospheric temperature and the downward-directed solar radiation fluxes over the Atlantic Ocean. Their model shows a decrease of the radiation flux reaching the surface of the earth and an increase of the tropospheric temperature which can be about l-2 K in the daily mean. Gillette (1981) describes a mechanism that can be considered responsible for the mobilization and pro- duction of those mineral dust particles which may be carried long distances. When the wind exceeds a certain threshold velocity coarse particles (radii > 20 pm) which lie loosely on the sand surface start to roll and saltate (soil aeepand saltation). Along their trajectory, the saltating sand particles gain momentum and once they strike the sand surface again the impact can be large enough to ‘splash’ agglomerates of bigger sand grains into smaller dust particles. Also. the Mating grain itself might break down into smaller particles after collision. By these impactions dust particles may be released from the ground and lifted up by turbulent diflusion processes finally entering into the long-range transport. Owen (1964) derives equations of motion for the trajectories of saltating sand particles and gives vel- ocity ranges in terms of the shear velocity in which saltation occurs. Saltating sand grains most effective for mobilization of mineral dust particles have radii of about 50 pm (Chepil, 1951). In a wind tunnel study Gillette (1978a) measured vertical and horizontal particle fluxes of ‘coarse’ sand particles (radii > 12.5 pm) and ‘fine’ sand particles (radii c 12.5 pm). The category of ‘fine particles’ has not been further differentiated with respect to particle size. Other wind tunnel experiments were performed by Gillette (1978b) to determine the threshold velocity over pebble-covered desert soils and a sandy agricul- tural soil. Knotterus (1971) studied the influence of the air moisture on the threshold velocity of dune sands by use of a wind tunnel. Fairchild and Tillery (1982) determined resuspension rates and vertical particle fluxes of aluminum spheres (count median diameter < 10 pm) that were loosely spread over a resuspension bed in a wind tunnel as a function of the size of the saltating glass beads. The objectives of this study are: (I) Establishing the mobilization mechanism (Gillete, 1981) in a suitably designed wind tunnel. (2) Measurement of aerosol particle size distri- butions in the turbulent boundary layer as a function of the shear velocity D*, the height above the sand surface and the rate or amount of saltating particles present in the flow. 1891